SECTION 1: Methods and techniques of OCT angiography examination
- CHAPTER 1: Principles of OCT angiographyYali Jia, Tristan T Hormel, David Huang
- CHAPTER 2: Interpretation of OCT angiographyTristan T Hormel, Yali Jia, David Huang
- CHAPTER 3: OCT angiography: TerminologyDavid Huang, Tristan Hormel, Yali Jia
- CHAPTER 4: OCT angiography in everyday clinical practiceBruno Lumbroso, Marco Rispoli, Maria Cristina Savastano
- CHAPTER 5: Retinal normal vascularizationMaria Cristina Savastano, Marco Rispoli, Bruno Lumbroso
- CHAPTER 6: Corneal and anterior segment OCT angiography
ABSTRACT
Optical coherence tomography angiography (OCTA) data is generated by measuring motion contrast between sequential optical coherence tomography (OCT) scans. Here we review how the OCT data is collected and how flow signal can be measured using either amplitude, phase, or complex signals.
INTRODUCTION
Optical coherence tomography (OCT) uses interferometry to measure tissue reflectance.1 Interferometry relies on the interaction between a reference beam and light reflected from the sample arm after interaction with the tissue. The depths of tissue reflections are resolved by coherence gating, which refers to the mutual coherence between reference and sample reflections. Transverse scanning of the beam in the sample arm makes OCT a three-dimensional imaging modality. OCT usually employs invisible infrared light, which is advantageous for patient comfort in ophthalmic applications. The axial resolution of OCT systems ranges from 2 to 10 µm, depending on the spectral bandwidth and wavelength.2 This enables noninvasive visualization of the internal layers of thin structures such as the retina not possible with any other technology. These advantages have made OCT the most commonly performed imaging procedure in ophthalmology,3 where it is used to diagnose disease and assess treatment efficacy.4
In structural OCT, inherent variation in tissue reflectivity enables the identification of different structures. For instance, the inner nuclear layer of the retina has relatively low reflectivity, and can be distinguished from the more reflective inner and outer plexiform layers around it. However, this does not provide good contrast for capillaries, which usually have similar reflectivity to the tissues in which they are embedded. Structural OCT measurements are consequently incapable of achieving adequate detail to construct an angiogram at capillary-scale detail. Early attempts at OCT angiography (OCTA) uses the Doppler shift measured between adjacent axial scans, but this proved unreliable because the OCT beam often strikes retinal blood vessel at near perpendicular incidence, which makes the Doppler shift too small to measure.5,6 Reliable OCTA eventually emerged as more robust methods to detect motion between successive OCT cross-sectional scans (B-scans) were developed.
GENERATING OCTA DATA FROM MOTION CONTRAST
Optical coherence tomography angiography relies on motion contrast to highlight blood vessels down to the capillary level. Blood flow changes the OCT reflectance signal between sequential B-scans (Figures 1A to D). This change constitutes flow signal.
Because OCTA is based on OCT data, it has many of the characteristics of structural OCT imaging. OCTA is also a noninvasive, three-dimensional modality. OCTA data is automatically coregistered with the structural OCT data used to produce it. This can be useful for assessing the location of vasculature relative to the tissue in which it is embedded and for correlating the structural and vascular features to enhance the diagnosis of retinal pathologies.
METHODS FOR MEASURING MOTION CONTRAST
There are a number of ways to measure motion contrast. OCT signal is complex valued—including both amplitude and phase components. Consequently, OCTA can be either phase-based, amplitude-based, or complex-signal-based.
The first attempts to achieve angiography from OCT devices relied on Doppler phase shifts. Doppler OCT can measure the absolute blood flow velocity based on the phase shift between consecutive axial scans and the beam incidence angle. The Doppler shift is proportional to the off-perpendicular angle between the OCT beam and the direction of blood flow. Unfortunately, for retinal OCT scanning, this angular offset is often close to zero. To overcome this limitation, researchers next turned to phase variance (rather than phase shift) as the flow signal.7–9 However, phase-based OCTA is very susceptible to corruption by phase noise due to bulk tissue motion and OCT system noise (especially swept-source output).10 There are several methods that compensate for phase noise, which largely rely on the statistical properties (e.g., the mean or histogram) of the flow signal distribution within an OCTA volume.9–12 While no method can completely remove phase noise, a recent approach that relies on using the standard deviation of flow values within a line scan can reliably and efficiently compensate for phase shifts.12
To avoid the difficulties with phase noise, most commercial OCTA systems are amplitude-based. While amplitude-based OCTA lose some flow sensitivity compared to phase measurements, it is sensitive enough to measure capillary. Because it is less susceptible to noise from tissue bulk motion and other sources of phase variation, amplitude-based OCT is more reliable and easier to implement.
Optovue, Heidelberg, and Topcon instruments all rely on amplitude-based motion contrast. But the exact algorithm differs. Heidelberg OCTA measures the temporal amplitude distribution within a given voxel in order to estimate the probability that it belongs to static tissue or a vessel.13 To achieve enough information to adequately sample, these amplitude distributions require 4–7 consecutive B-frames at each scan location.13 Optovue systems employ the split-spectrum amplitude-decorrelation angiography (SSADA) algorithm that requires only two consecutive B-frames for compute a high-quality angiogram. Topcon instruments use a ratio analysis approach [termed “OCTA ratio analysis (OCTARA)”] in which the ratio between the minimum and maximum voxel value 4at two different time points is compared in order to construct the OCTA signal.14 This algorithm also requires at least four repeated B-scans to achieve adequate results.
Figures 1A to D: Optical coherence tomography angiography (OCTA) signal generation. (A) Two sequential cross-sectional structural OCT scans (scan A and scan B) are generated by collecting data from a sample beam at a detector. (B) When the sample beam (red and blue arrows) encounters a blood vessel, flowing blood imparts a change in the reflectance signal between scan A and scan B. (C) On the other hand, when the sample beam encounters static tissue, the reflectance signal in scan A and scan B will be essentially identical. (D) By measuring the change between scan A and scan B, blood flow can be identified.
Complex-signal-based OCTA uses both the phase and amplitude components of the OCT signal. It is highly susceptible to phase noise like phase-based OCTA. The most prominent complex-signal-based OCTA generation algorithm is “optical microangiography” (OMAG). This algorithm uses frequency modulation in the interferogram in order to separate the static signal from the flow signal. The specifics of how this offset is achieved have changed as the technique improved over time.15–18 OMAG requires adequate phase compensation in order to remove noise from bulk motion. Zeiss instruments use the ultrahigh sensitive OMAG algorithm,19 which has recently achieved high-quality angiographic images from just two repeated B-scans on 100-kHz swept-source OCT prototype (Figures 2A to F).20,21
SPECTRAL SPLITTING
Optical coherence tomography phase, amplitude, and complex signals can all be enhanced using spectral splitting, in which the OCT signal is processed separately in different frequency sub-bands and then averaged to produce the OCTA angiogram. Spectral splitting improves the flow detection signal-to-noise ratio (SNR) (and, consequently, downstream measurements such as vessel density or connectivity). The enhanced signal comes at the cost of reduced axial resolution, since each of the constituent frequency bands must be narrower than the full spectrum (which achieves optimal resolution). In ophthalmic imaging, this is not problematic since even at reduced axial resolution, spectrally split OCTA measurements can still unambiguously separate the vascular plexuses. Lowering axial resolution by spectral splitting reduces susceptibility to noise due to cardiac pulsation and other axial bulk motion, which further enhances the SNR of flow detection.
The first commercial OCTA instruments were developed by Optovue and made use of SSADA.22 SSADA is a purely amplitude-based OCTA processing algorithm, but research studies have made use of spectral splitting for phase- and complex-based processing as well.23 In each case, improvements in image SNR and contrast have been measured. Optovue instruments employing the SSADA algorithm require just two repeated B-scans in order to construct OCTA volumes.24 Due to this efficiency, recently SSADA has been able to achieve 12 × 12-mm field of view in a single scan (Figure 3) using the latest 120-kHz Solix system (Optovue, Inc.).
CONCLUSION
Optical coherence tomography angiography uses motion contrast to detect flow down to the capillary level. Flow signal is computed from the change in OCT reflectance between consecutive B-scans. Several different approaches can be used to compute the flow signal. The most efficient algorithms can obtain adequate flow SNR and image quality using only two consecutive B-scans at each location.5
Figures 2A to F: Ultrahigh sensitive optical microangiography (OMAG) images of retinal vasculature at several scales compared to fundus photography. (A) A fundus photography image of a healthy retina. (B) A montaged OMAG en face image of the nerve fiber layer. (C) A superficial retinal slab shows the vascular network in the ganglion cell layer and outer plexiform layer. (D) Retinal slab corresponding to the deep vascular complex. (E) Image showing the vasculature in both (C) and (D), with vessels color-coded according to depth (red: superficial; green: intermediate; blue: deep). (F) Magnified image detailing the blue box in (E), along with a structural cross-section with flow overlaid at the position indicated by the dotted dashed line. Capillary details are clearly visible in the magnified version.Source: Reprinted with permission from Zhang Q, Lee CS, Chao J, et al. Wide-field optical coherence tomography based microangiography for retinal imaging. Sci Rep 2016; 6:22017.
Figure 3: 12 × 12-mm, 600 × 600-pixel resolution image of a normal retina from a commercial instrument (Solix, Optovue, CA) employing SSADA OCTA processing. This efficient algorithm requires only two sequential scans to capture the detail shown here.
REFERENCES
- Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science 1991;254:1178–1181.
- Drexler W, Fujimoto JG. Optical coherence tomography: technology and applications. Switzerland: Springer International Publishing, 2010.
- Swanson EA, Huang D. Ophthalmic OCT reaches $1 billion per year. Retin Physician 2011;45:58–59.
- Brand CS. Management of retinal vascular diseases: a patient-centric approach. Eye 2012;26:S1–S16.
- Chen Z, Milner TE, Srinivas S, et al. Noninvasive imaging of in vivo blood flow velocity using optical Doppler tomography. Opt Lett 1997;22:1119–1121.
- Izatt JA, Kulkarni MD, Yazdanfar S, Barton JK, Welch AJ. In vivo bidirectional color Doppler flow imaging of picoliter blood volumes using optical coherence tomography. Opt Lett 1997;22:1439–1441.
- Fingler J, Schwartz D, Yang C, Fraser SE. Mobility and transverse flow visualization using phase variance contrast with spectral domain optical coherence tomography. Opt Express 2007;15:12636–12653.
- Fingler J, Readhead C, Schwartz DM, Fraser SE. Phase-contrast OCT imaging of transverse flows in the mouse retina and choroid. Invest Ophthalmol Vis Sci 2008;49:5055–5059.
- Makita S, Hong Y, Yamanari M, Yatagai T, Yasuno Y. Optical coherence angiography. Opt Express 2006;14:7821–7840.
- Szkulmowski M, Grulkowski I, Szlag D, Szkulmowska A, Kowalczyk A, Wojtkowski M. Flow velocity estimation by complex ambiguity free joint spectral and time domain optical coherence tomography. Opt Express 2009;17:14281–14297.
- Wang RK, An L. Doppler optical micro-angiography for volumetric imaging of vascular perfusion in vivo. Opt Express 2009;17:8926–8940.
- Wei X, Camino A, Pi S, et al. Fast and robust standard-deviation-based method for bulk motion compensation in phase-based functional OCT. Opt Lett 2018;43:2204–2207.
- Rocholz R, Teussink MM, Dolz-Marco R, et al. SPECTRALIS optical coherence tomography angiography (OCTA): principles and clinical applications. Heidelb Eng Acad 2018; 1–10.
- Reisman CA, Wang Z, Liu JJ, et al. Swept source OCT angiography based on ratio analysis. Invest Ophthalmol Vis Sci 2016;57:452.
- Wang RK, Jacques SL, Ma Z, et al. Three dimensional optical angiography. Opt Express 2007;15:4083–4097.
- An L, Wang RK. In vivo volumetric imaging of vascular perfusion within human retina and choroids with optical micro-angiography. Opt Express 2008;16:11438–11452.
- An L, Qin J, Wang RK. Ultrahigh sensitive optical microangiography for in vivo imaging of microcirculations within human skin tissue beds. Opt Express 2010;18:8220–8228.
- An L, Shen TT, Wang RK. Using ultrahigh sensitive optical microangiography to achieve comprehensive depth-resolved microvasculature mapping for human retina. J Biomed Opt 2011;16:106013.
- Tan AC, Tan GS, Denniston AK, et al. An overview of the clinical applications of optical coherence tomography angiography. Eye (Lond) 2018;32:262–286.
- Wang RK, Zhang A, Choi WJ, et al. Wide-field OCT angiography enabled by 2 repeated measurements of B-scans. Opt Lett 2016;41:2330–2333.
- Zhang Q, Lee CS, Chao J, et al. Wide-field optical coherence tomography based microangiography for retinal imaging. Sci Rep 2016;6:22017.
- Jia Y, Tan O, Tokayer J, et al. Split-spectrum amplitude-decorrelation angiography with optical coherence tomography. Opt Express 2012;20:4710–4725.
- Liu G, Jia Y, Pechauer AD, Chandwani R, Huang D. Split-spectrum phase-gradient optical coherence tomography angiography. Biomed Opt Express 2016;7:2943–2954.
- Gao SS, Liu G, Huang D, Jia Y. Optimization of the split-spectrum amplitude-decorrelation angiography algorithm on a spectral optical coherence tomography system. Opt Lett 2015;40:2305–2308.